“I've just been bitten by a dog”

BMJ 1997; 314 doi: https://doi.org/10.1136/bmj.314.7074.88 (Published 11 January 1997) Cite this as: BMJ 1997;314:88

Surgical toilet, appropriate antibiotics, and advice to come back if infection develops

  1. Fionna Moore, Consultant, accident and emergency servicea
  1. a Charing Cross Hospital, London W6 8RF

    Mammalian bites have a sinister reputation for causing tissue damage and infection. Although human bites have the greatest potential for local injury–because of the varied and virulent flora of pathogenic organisms in the mouth and the propensity for association with a crush injury–dog bites are numerically the most common.1 They account for 1-2 million injuries in America each year2 and about 200 000 cases in Britain.3 In addition, many victims regard their injury as too insignificant to seek medical help initially. A small number may then go on to develop overwhelming systemic infection, as reported in this week's BMJ by Mellor et al (p 129).4 This suggests that dog bites can only be regarded as trivial in retrospect. What then should be the advice to patients, general practitioners, minor injuries units, and accident and emergency departments in terms of safe guidelines for management?

    The treatment of dog bites is twofold: proper surgical toilet and appropriate antibiotic treatment where indicated. Wound toilet remains the mainstay of treatment of all bite injuries, whether this means adequate cleaning of a superficial wound by the patient at home or the full exploration, debridement, and irrigation of a more extensive injury under local, regional, or general anaesthesia. The question of whether to close the wound depends on the age, site, and nature of the injury. Many bite wounds include a substantial crush injury and potentially heavy contamination with organisms from the dog's mouth. These should be closed by delayed primary suture. Wounds of the face, including the scalp and ear, where the blood supply is excellent and the incidence of infection low, can safely be closed primarily after thorough cleaning and removal of any dead or devitalised tissue. Wounds elsewhere should be carefully assessed on the basis of time elapsed since the bite and the extent of crush injury. When the wound is recent–less than six to eight hours old–and the degree of crush injury is minimal–as elicited from a careful history of the mechanism of the bite–it may be safe to consider primary closure after thorough cleaning and irrigation. When the wound is older and clearly contused with a major crush element, delayed primary closure is the treatment of choice.

    What of the role of antibiotics? Many accident and emergency departments prescribe prophylactic antibiotics routinely for all bites. However, American studies have suggested that, although antibiotics reduce the incidence of infection, 14 patients may have to be treated to prevent one infection.5 Given the cost implications and the likely poor compliance of many patients, this may not be justified. An American study in 1996, which used careful exclusion criteria, suggested that low risk wounds from dog bites carried no greater risk of infection if prophylactic antibiotics were not used, though proper wound toilet was emphasised.6 It is possible to identify high risk wounds and patients that indicate use of prophylactic antibiotics no matter how apparently insignificant the injury. For wounds, the following are clearly high risk injuries: wounds more than eight hours old, crush or puncture wounds, and wounds to the hands or feet. Wounds which have been closed primarily should, in my view, be covered with prophylactic antibiotics. However, this should clearly be in addition to, not instead of, thorough wound toilet. For patients, high risk is associated with age over 50 years, female sex, alcoholic liver disease, asplenism, immunosuppression, and immunological compromise.

    If antibiotics are prescribed the most appropriate is co-amoxiclav 500/125 (375 mg) three times daily for five days. For patients who are allergic to penicillin, doxycycline 200 mg daily is suggested, or erythromycin for children under the age of 12 and pregnant or breast feeding mothers.7 These regimens will cover the most common infecting organisms – Pasteurella multocida, streptococcal species, anaerobes, and Staphylococcus aureus–as well as other common commensals of the dog's mouth, such as Capnocytophaga canimorsus, which rarely cause infection. Erythromycin is the least effective drug. All patients, however they are treated or advised, should be warned of the signs of developing infection and told who to contact for further assessment.

    Would adherence to such guidelines prevent cases such as that described in this journal? A handful of such cases, relating to infection with Capnocytophaga canimorsus, have been described in the past five years.8 9 10 This is a well recognised risk in asplenic or immunocompromised patients,11 12 but other cases have occurred in middle aged or elderly victims who were otherwise in good health (Garrard C, personal communication). In the cases described, however, the patients would have fallen into high risk groups if medical advice had been sought. The organism is slow growing and not always easy to culture, and several cases have initially been diagnosed clinically as meningococcaemia, so the real incidence may be higher.

    A patient's tetanus immunisation status should also be checked and updated if necessary. Given the much greater general access to foreign travel, patients should be reminded that rabies continues to flourish beyond our shores and that they should seek advice if they sustain an animal bite or lick to an open wound when abroad.

    In summary, patients should be made aware of the potentially harmful effects of even apparently trivial injuries, particularly if they are dog owners in high risk groups, and so should their medical and nursing attendants. While the importance of proper treatment of minor or moderate dog bites in terms of proper public and professional education should be highlighted, it may be appropriate to consider the problem of potentially life threatening injuries from dog bites. Two further cases were recently reported in the national press (Daily Telegraph, 30 November 1996) involving two babies, of three days and two weeks of age respectively. Children, particularly those under the age of 5 years, are particularly at risk,13 especially with injuries to the head and neck, and the perpetrators are often medium sized or large dogs familiar to the family. Targeting patients with information about the potential problems arising from apparently trivial bite injuries should perhaps be linked to the possible prevention of more serious injury to a section of the population who are least able to protect themselves from the unwanted attention of “man's best friend.”


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    7. 7.
    8. 8.
    9. 9.
    10. 10.
    11. 11.
    12. 12.
    13. 13.
    View Abstract